[open-science] Why are we leaving it to Google?

[Bosman, J.M. (Jeroen)]

Some observations:

1) Google is hardly the first, we've already got:
- Elsevier datasearch (commercially owned, free for now, not open) https://datasearch.elsevier.com/#/
- Datacite search (non-profit, community governed, with API, free but not entirely open): https://search.datacite.org
- Worldwidescience dataset filter (publicly owned, free, not open): https://worldwidescience.org/
- Data Citation Index from Clarivate (commercially owned, costly, closed): https://clarivate.com/products/web-of-science/web-science-form/data-citation-index/
and also some (many?) field sepecific data search engines, such as:
- BioCaddie/Datamed (publicly owned, free): https://biocaddie.org/  / https://datamed.org/
- Neuroscience Information Framework (publicly owned, free): https://neuinfo.org/data/search?q=*&l=#all

2) All of these search engines have their strengths and weaknesses, particularly due to limited and varying metadata. Hope to do a more thorough assessment of these and others sometime.

3) Google Scholar probably is not the largest scholarly search engine anymore. That is quite possibly LENS, built on the (open) data from MAS and Pubmed.

But this is getting a bit off topic. My point is we are luckily not leaving it entirely up to Google. Also we should separate the quality/functionality of the Google dataset search (it is probably too early to assess that) from the problems deriving from the fact that it is closed, lacks an API etc.. And yes, it would be great to have an entirely open system for dataset search next to these existing ones.

Jeroen Bosman
Utrecht University Library
________________________________________
From: open-science [open-science-bounces at lists.okfn.org] on behalf of Peter Kraker [pkraker at openknowledgemaps.org]
Sent: Monday, September 10, 2018 7:01 PM
To: Thomas Krichel; Emanuil Tolev
Cc: open-science; opencon-discussion-list at googlegroups.com
Subject: Re: [open-science] Why are we leaving it to Google?

I agree that DOAJ is great and has a great search engine. We even used
it as our default data source for Open Knowledge Maps for some time.
However, it lacks coverage. Researchers will always choose the search
engine with the largest (perceived) coverage, and that's Google Scholar.
And no one has the resources to beat them in that area. So as long as
Google doesn't allow others to reuse their index, no real innovation in
scholarly discovery will happen.

Aren't you upset by the fact that in every other area you can easily get
an overview of what exists and find exactly what you are looking for
(websites, products etc), but when it comes to research, you are left
with huge piles of papers with little to no context?

And now they are going to do the exact same thing with datasets. I don't
think that we should let that happen.

Best,
Peter

On 10/09/2018 18:39, Thomas Krichel wrote:
> Emanuil Tolev writes
>
>> Well, there's value in having people funded to think about those things
>> independently on medium and long term. That's why https://doaj.org exists.
> I am not disputing the value of DOAJ. DOAJ is a dataset. I'm disputing
> the value of putting resources into building a search engine when
> the same resources could be put into building datasets, and when
> Google already are building a search engine.
>
>> It probably provides a better search experience in some ways and worse in
>> other ways/contexts. Either way the org behind it is very useful and the
>> people on the editorial team have gathered quite a lot of interesting
>> knowledge that orgs like SCOSS and SPARC, foundations, funders etc. then
>> make use of.
> I agree.
>
>> Google eng wouldn't think of the problems in the same way as
>> the open science people involved in the open* mailing lists and DOAJ.
> I'm not sure what Google "eng" is but if you mean engineers, I for
> one would not stereotype them in this way.
>
> BTW, I would be grateful if a maintainer of opencon could add me to
> that group.
>

_______________________________________________
open-science mailing list
open-science at lists.okfn.org
https://lists.okfn.org/mailman/listinfo/open-science
Unsubscribe: https://lists.okfn.org/mailman/options/open-science
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-science/attachments/20180910/77c2bb5b/attachment-0002.html>